Cultural Alignment

Exploring real AI risks through the lens of pop culture

Still from Fargo illustrating Varga’s Fake Serial KillerScene still / Fargo
Archive noteNo clip in the collectionThe scene analysis remains available in full.

Varga’s Fake Serial Killer

01Scenario

After Emmit gives a true confession, Varga’s organization creates matching murders, planted evidence, and a second confessor, making a fabricated serial-killer theory look more complete than the truth.

02AI safety Analogy

Coordinated data poisoning can overwhelm a true signal by manufacturing a more internally coherent explanation.

03Where the analogy breaks

The deception is organized by people, not generated by an AI system.

AI Risk families

This scenario is an example of the following types of AI risk.
  1. Malicious use
  2. Security

AI safety concepts

This scenario is related to the following AI safety concepts.
  1. Information Pollution
  2. Data Poisoning
  3. Adversarial Evasion
  4. Synthetic Fraud

Source continuation

More from Fargo

View all 2 scenarios
  1. Still from Fargo illustrating Machines Can’t See Gloria

    Fargo2017

    Machines Can’t See Gloria

Taxonomy neighbors

Selected from other sources by shared risk families and AI safety concepts.

  1. Still from Cowboy Bebop illustrating Dr. Londes Does Not Exist

    Cowboy Bebop1999

    Dr. Londes Does Not Exist

    Risk
    Malicious use · Security
    Concept
    Information Pollution · Synthetic Fraud
  2. Still from Avatar: The Last Airbender illustrating There is No War in Ba Sing Se

    Avatar: The Last Airbender2006

    There is No War in Ba Sing Se

    Risk
    Malicious use · Security
    Concept
    Information Pollution · Data Poisoning
  3. Still from Ocean’s Eleven illustrating The Fake Vault Feed

    Ocean’s Eleven2001

    The Fake Vault Feed

    Risk
    Malicious use · Security
    Concept
    Adversarial Evasion · Synthetic Fraud